Hyaluronic acid ameliorates the proliferative ability of human amniotic epithelial cells through activation of TGF-β/BMP signaling

Reagents

Dulbecco’s modified Eagle medium with low glucose (LG-DMEM), F-12 medium, trypsin, GlutaMAX, non-essential amino acids, fetal bovine serum (FBS), and adipocyte/osteocyte/chondrocyte differentiation basal medium and their corresponding supplements were purchased from Gibco (NY, USA). Antibiotics (penicillin and streptomycin) were purchased from Sino American Biotechnology (Shanghai, China). HA was obtained from Seebio Biotech, Inc. (Shanghai, China). Bovine serum albumin (BSA), hematoxylin, saturated oil red O dye solution, Marker I DNA Ladder, and Gold View Type I nucleic acid staining were obtained from Solarbio (Beijing, China). Alizarin red dye solution and Toluidine blue were purchased from Cyagen (Shanghai, China) and Sangon Biotech (Shanghai, China), respectively. SB431542 was obtained from Selleck (Shanghai, China). Cell Counting Kit-8 (CCK-8) was purchased from Dojindo (Beijing, China). Human IL-10 ELISA Kit and human TGF-β1 ELISA Kit were purchased from Biosamite (Shanghai, China). All other chemicals were of analytical grade.

Ethics

All amniotic membranes were collected from healthy donors undergoing caesarean delivery and the informed written consent was acquired from the donor prior to tissue collection. The study protocol and the use of the human amniotic membrane were approved by the Animal Experiment Ethics Committee of Zunyi Medical University (Zunyi, China) (License number granted: (2014)2-085) and Medical Ethics Committee of Zunyi Medical University (License number granted: (2014)1-042) and the declaration of Helsinki was strictly abode by the investigators.

Isolation and cultivation of hAECs

hAECs were obtained by mechanical separation and the following enzymatic digestion. Briefly, the amniotic membranes were mechanically separated from the chorion of fresh term placentae. After completely washing away the blood with D-Hank’s solution containing penicillin (100 U/mL) and streptomycin (100 µg/mL), the amnions were minced and transferred into a 50-mL tube containing 0.05% trypsin and 0.02% EDTA solution. Then, the tube was maintained at 37 °C for 40 min in a water bath shaker at 180 rpm. The digested tissue pieces were filtered through a 300 mesh stainless steel sieve. The cell suspension was collected by centrifugation and resuspended in LG-DMEM/F12 complete medium (containing 10% FBS) to obtain primary hAECs (passage 0, P0). Cells were inoculated into T25 flasks at a density of 2. 5 ×10⁶/flask and were cultured at 37 °C in a 5% CO₂ incubator, and the medium was replaced once every 72 h. Cells were passaged when they reached 80% confluence. All of the following assays repeated at least for 3 times and each experimental replicate used a batch of hAECs derived from a different placental donor. Cells from multiple batches were not mixed together and cultured.

Phenotypic properties of hAECs

The phenotype of hAECs was validated by flow cytometric analysis in Affiliated Hospital of Zunyi Medical University. Cells were incubated with antibodies (mouse anti human, BD, NJ, USA) against CD44 (Cat# 555478) and CD45 (Cat# 561865), conjugated to fluorescein isothiocyanate, antibodies against CD34 (Cat# 550619), CD29 (Cat# 557332), CD73 (Cat# 550257), CD166 (Cat# 559263), conjugated to phycoerythrin, and the antibody against HLA-DR (Cat# 552764) conjugated to PerCP-Cy5.5, at room temperature for 25 min in dark, and analyzed by a FACSCalibur system (BD, Franklin Lakes, NJ, USA) using CellQuest software after PBS washing. Non-immune isotype IgG antibodies (BD, NJ, USA) were used as controls. Similar to our previous study (Luo et al., 2019), hAECs highly expressed the surface markers such as CD29 (99.62%), CD73 (99.60%), and CD166 (99.39%), and lowly expressed CD44 (17.6%), but did not express the hematopoietic stem cells marker CD34 (0.09%) and CD45 (1.17%), and the MHC II molecule HLA-DR (0.02%). The absence of CD34 and CD45 positive cells indicated that the isolates were not contaminated with hematopoietic stem cells from umbilical cord blood or embryonic fibroblasts.

Immunocytochemistry

The first generation (passage 1, P1) hAECs in logarithmic growth phase were seeded in 6-well plates at a density of 3. 0 ×10⁵/well, and cytokeratin 19 (CK19) and vimentin were measured by immunocytochemical staining. In brief, cells were rinsed with D-PBS and fixed in 4% paraformaldehyde for 30 min at 24−28 °C. Cells were then treated with 0.5% Triton-X 100 for 10 min and 5% BSA for 30 min at 24−28 °C, followed by incubation with antibodies (mouse anti human, 1:100, Gene Tech, Chengdu, China) against CK19 (Cat# GM088804) or vimentin (Cat# GM072504) overnight at 4 °C. Cells were then washed and incubated with horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG secondary antibody for 30 min at 37 °C. Finally, the labeled cells were stained using a DAB kit (Gene Tech, Chengdu, China) and observed under a microscope after washing. Nuclei were counterstained with hematoxylin. Cells incubated with PBS were used as controls.

Proliferation assay

The P1 hAECs in the logarithmic growth phase were seeded in 96-well plates at densities of 1, 2, 4, 8, 16, 32 and 64 ×10³/well, respectively. After incubation at 37 °C for 6 h, cells were treated with CCK-8 kit according to the manufacturer’s instruction. The absorbance of the cells in each well was measured at 450 nm using a microplate reader (MultiskanTM GO, ThermoFisher, USA), and a standard curve was subsequently generated and used to calculate the cell numbers. To explore the effect of HA with different molecular weights on cell proliferation, the P1 hAECs in logarithmic growth phase were seeded in a 96-well plate at a density of 1. 0 ×10⁴/well, the medium was replaced on day 3 after inoculation. The cells were then treated for 48 h by 0.5 mg/mL of HA with different molecular weights including 50 kDa, 300 kDa, and 1,000 kDa. Finally, the proliferation capability was measured by CCK-8 assay following the manufacturer’s instructions. To evaluate the effect of 300 kDa HA at different doses on cell proliferation, the P1 hAECs in logarithmic growth phase were seeded in a 96-well plate at a density of 1. 0 ×10⁴ cells/well, the medium was replaced on the day 3 after inoculation. The cells were then treated for 48 h with 300 kDa HA at different doses of 0.05, 0.1, 0.5, and 1 mg/mL. The proliferation capability was measured by CCK-8 assay. Cell doubling time (DT) was calculated by the following formula (Zhou et al., 2013): DT = t × log2/(logN_t−logN₀), where t was treatment time with HA, N₀ was the seeded cell number, N_t was the harvested cell number after HA treatment for t hours. In this case, t was 48 h.

in vitro differentiation potentials

P1 hAECs in logarithmic growth phase were inoculated into 6-well plates at a density of 3. 6 ×10⁵/well. hAECs in the presence and absence of HA (300 kDa, 1 mg/mL) were incubated in LG-DMEM/F12 complete medium at 37 °C for 3 days, and the medium was subsequently replaced with differentiation complete media for adipocyte, osteocyte, and chondrocyte, respectively. Adipogenic differentiation complete medium was composed of LG-DMEM supplemented with Dex (1 µM), indomethacin (200 µM), IBMX (500 µM), and recombinant human insulin (20 mg/mL). Osteogenic differentiation complete medium was composed of LG-DMEM supplemented with Dex (0.1 µM), β-glycerol phosphate (5 mM) and ascorbic acid 2-phosphate (50 µg/mL). Chondrogenic differentiation complete medium was composed of LG-DMEM supplemented with Dex (0.1 µM), ascorbic acid 2-phosphate (50 µg/mL), TGF-β3 (10 ng/mL) and 1% ITS. These differentiation media were replaced every 3 days. After induction, differentiation properties were assayed on the 14th day for adipogenesis, and on the 21st day for osteogenesis and chondrogenesis. Briefly, cells were fixed using 4% paraformaldehyde for 30 min and washed with PBS before staining. For adipogenic staining, cells were stained with oil red O staining solution (saturated oil red O solution: ddH₂O = 3:2, then filtered with neutral filter paper). For osteogenic staining, cells were stained by Alizarin red. For chondrogenic staining, cells were stained with Toluidine blue solution.

Immunofluorescence assay

P1 hAECs in logarithmic growth phase were inoculated into 6-well plates at a density of 3 ×10⁵/well, and cultured at 37 °C with 5% CO₂ for three days in LG-DMEM/F12 complete medium. The cells were further cultured in LG-DMEM/F12 complete medium with or without HA (300 kDa, 1 mg/mL) for two days. Immunofluorescence staining was then performed to detect the expression of stemness-related proteins (Nanog, Sox-2 and Oct-4). Briefly, after fixation, washing, and blocking steps, as did in the immunocytochemistry assay, the cells were incubated with rabbit-derived primary antibodies (Abcam, Cambridge, UK) against Nanog (1:200; Cat# ab109250), Sox-2 (1:100; Cat# ab137385), and Oct-4 (1:250; Cat# ab200834) at 4 °C overnight. The cells were then washed by PBS thrice and incubated with a FITC-conjugated goat anti-rabbit IgG secondary antibody (1:32, Abcam, Cambridge, UK) for 2 h at 24−28 °C. The cells were washed thrice again to remove the secondary antibody and then counterstained with 4′,6-diamidino-2-phenylindole (DAPI, Roche, Basel, Switzerland). Cells were then observed under a fluorescent microscope (Leica DMIRB, DIC, Germany). Goat serum was used as a negative control.

Enzyme-linked immunosorbent assay

Enzyme-linked immunosorbent assay (ELISA) was performed according to the manufacturer’s instructions. After the cell supernatants of each group were collected at 24−28 °C, the target proteins were concentrated by centrifugation for 10 min at 4,000 g. The supernatant of cells was concentrated to 1/12 of its original volume. The ELISA kit was equilibrated at 24−28 °C for 20 min, followed by the addition of 50 µL of different concentrations of standards to each standard well, and 10 µL of the sample to be tested and 40 µL of the sample diluents to each testing well. One hundred microliters of horseradish peroxidase (HRP)-labeled detection antibody were added to each standard well and sample well. Then the reaction well was sealed and incubated at 37 °C for 60 min. After repeated washing, substrates A and B (each 50 µL) were successively added to each well for color development. The samples were incubated at 37 °C in the dark for 15 min. Finally, 50 µL of stop solution was added to each well and the OD value was measured at 450 nm within 15 min using a platereader (MultiskanTM GO, ThermoFisher, USA).

Cellular senescence assay

To evaluate the effect of HA on cellular senescence, P1 hAECs in logarithmic growth phase were inoculated into 6-well plates at a density of 3 ×10⁵/well and P2 hAECs were cultured in the presence or absence of HA (300 kDa, 1 mg/mL) for 48 h. After cell passage, cellular senescence of P3 hAECs was detected using β-Galactosidase Staining Kit (Beyotime, Shanghai, China) following the manufacturer’s instructions. The stained cells were then observed under a microscope and cells from 9 random fields were selected to count the rate of SA-β-Gal positive cells to total cells for each group.

RT-qPCR analysis

RNA was extracted from all samples using the RNAiso Plus Kit (TaKaRa, Dalian, China) according to the manufacturer’s protocol. The concentration and purity of extracted RNA were measured using the NanoDrop 2000c (Thermo Scientific, Bonn, Germany) at OD260 and OD260/280, respectively. Reverse transcription was conducted using the PrimeScript™ RT reagent Kit (TaKaRa, Dalian, China) according to the manufacturer’s instructions. PCR was carried out using the Premix Ex TaqTM II (for Real Time) (TaKaRa, Dalian, China) on a CFX96 Touch real-time PCR detection system (Bio-Rad, Hercules, CA, USA). The primers used in the reaction are shown in Table 1. The 2^−ΔΔCt method was used to quantify gene expression levels. β-Actin was used as the internal control for normalization.

PCR microarray

The potential signaling pathways by which HA affected the proliferation of hAECs was preliminarily screened by a PCR microarray. P1 hAECs in logarithmic growth phase were harvested, inoculated into 6-well plates at 3 ×10⁵/well, and cultured with 5% CO₂ at 37 °C. The medium was changed after 3 days, and then the cells were cultured for 18 or 36 h in the presence and absence of HA (300 kDa, 1 mg/mL). The total RNA was purified using the RNeasy MinElute kit (Qiagen, Valencia, CA, USA) according to the manufacturer’s instructions. cDNA was synthesized using a SuperScript III Reverse Transcriptase Kit (Invitrogen, Shanghai, China). Real-time PCR was conducted, and the 2^−ΔΔCt method was used to quantify gene expression levels. The signal was collected by Agilent Feature Extraction software and analyzed by Agilent GeneSpring GX software.

Validation of the signaling pathway

P1 hAECs in logarithmic growth phase were harvested, inoculated into 6-well plates at 3 ×10⁵/well, and cultured with 5% CO₂ at 37 °C. The medium was changed after 3 days, and the cells were then treated for 48 h in the presence or absence of HA (300 kDa, 1 mg/mL). The expression of differential genes related to the TGF-β/BMP signaling pathway revealed by the PCR microarray was validated using RT-PCR. To further unravel the direct correlation between the HA-induced cell proliferation and the TGF-β/BMP signaling pathway, we added SB431542 (10 µM), a specific blocker of the TGF-β/BMP signaling pathway, to the HA (300 kDa, 1 mg/mL)-treated group. Cell proliferation reflected by OD values were measured using the CCK-8 kit after 48 h of treatment. At the same time, the proliferation-related and TGF-β/BMP signaling pathway-related genes were detected by RT-PCR. Expression levels of proteins Smad2/3 and p-Smad2/3 were measured by Western Blot. The primary antibodies used were as follows: rabbit anti-human Smad2/3 (1:1000; Cat# 8828S, Cell Signaling Technology, Boston, USA), rabbit anti-human p-Smad2/3 (1:1000; Cat# 3102S, Cell Signaling Technology, Boston, USA), and rabbit anti-human GADPH (used as the reference protein) (1:1000; Cat# 10494-1-AP, ProteinTech Group, Chicago, USA). The secondary antibody used was HRP-labeled goat anti-rabbit IgG (1:3000; Cat# SA00001-2, ProteinTech Group, Chicago, USA). The protein bands were developed by ECL using Bio-Rad ChemiDocTM MP Imaging System, exposed using Image Lab software, and semi-quantified by ImagingJ software. The other three groups, including the negative control group treated without neither HA nor SB431542, the group treated with HA (300 kDa, 1 mg/mL) but without SB431542, and the group treated with SB431542 (10 µM) but without HA, were used as controls.

Cell cycle assay

P1 hAECs in logarithmic growth phase were plated into T25 culture flasks at 7. 81 ×10⁵/flask, and cultured in 5% CO₂ at 37 °C. The medium was changed after 3 days. Then, 300 kDa HA at different concentrations (0, 0.05, 0.1, 0.5, and 1 mg/mL) was added into the cells. After 48 h of treatment, the resultant cells were collected and stained with propidium iodide (PI) working solution according to the instruction of the Cell Cycle Kit (Beyotime, Shanghai, China). RNaseA had been included in the Kit to degrade RNA to ensure that only DNA was bound. The cell cycle was detected by flow cytometry (FACSCalibur, BD, Franklin Lakes, NJ, USA).

Statistical analysis

All experimental data were expressed as mean ± standard deviation (sd) and statistically analyzed using ANOVA test and Student’s t test by SPSS 19.0 software. The normal distribution of the data is verified by Agostino-Pearson omnibus normality test. Post-hoc comparisons were performed using Tukey’s multiple comparisons test. P <  0.05 was considered to be statistically significant.

Effects of HA on the proliferation of hAECs

HA is a kind of polysaccharide whose biological effects are greatly affected by its molecular weight (Cyphert, Trempus & Garantziotis, 2015). Thus, we firstly explored the effect of the molecular weight of HA on the proliferation of hAECs. The cells were exposed to 0.5 mg/mL HA with the molecular weights of 50, 300, and 1,000 kDa. After 48 h of treatment, compared with the control group without HA, the cell number in the 1,000 kDa and 50 kDa HA groups was decreased by 6.7% (P <  0.01) and 4.5% (P <  0.05), respectively, while the cell number in the 300 kDa HA group was increased by 7.4% (P <  0.01). These results indicate that 50 kDa and 1,000 kDa HA inhibited, yet 300 kDa HA promoted the proliferation of hAECs (Fig. 1A). The effect of 300 kDa HA at different concentrations on the proliferation of hAECs was further investigated (Fig. 1B). After 48 h of treatment, compared with the control group, the number of cells in the 0.05, 0.1, 0.5, and 1 mg/mL HA groups was increased by 7.6% (P <  0.01), 2.1% (P <  0.05), 8.8% (P <  0.01), and 10.5% (P <  0.01), respectively (Fig. 1C). Consistently, further study showed that the population doubling time (DT) of the cells was shortened from 58.47 h in the control group to 51.83 (P <  0.05), 56.37, 50.93 (P <  0.05) and 49.97 h (P <  0.05) in the 0.05, 0.1, 0.5, and 1 mg/mL HA groups, respectively (Fig. 1D). It is evident that 300 kDa HA could promote the proliferation of hAECs in the concentration range of 0.05–1 mg/mL, and the pro-proliferative effect of HA at 1 mg/mL was the most significant. We further measured the effect of 300 kDa HA on the expression of proliferation-associated genes Ki67 and PCNA after 48 h of treatment. HA at 0.5 and 1 mg/mL could boost the transcription of Ki67 and PCNA in hAECs, and in particular, HA at 1 mg/mL presented a significant increase as compared to the control group (P <  0.05) (Fig. 1E).

Effects of HA on the morphological and phenotypic characteristics of hAECs

As shown in Fig. 2A, the morphology of hAECs after 48 h of HA (300 kDa, 1 mg/mL) treatment was consistent with that of the control group, exhibiting ovoid or triangular shapes with a typical paving stone-like arrangement. Immunocytochemistry staining indicates that after HA exposure, hAECs still expressed the epithelial cell marker cytokeratin 19 (CK19), but not the mesenchymal cell marker vimentin (Fig. 2B). Flow cytometric analysis shows that after HA treatment, hAECs still expressed high levels of CD29, CD73 and CD166, low level of CD44, and did not express the hematopoietic stem cell surface markers such as CD34, CD45, and HLA-DR (Fig. 2C and Fig. S1). The above results indicate that 300 kDa HA treatment does not alter the morphological and phenotypic characteristics of hAECs.

Effects of HA on the stemness property and differentiation potential of hAECs

In comparison with the control, HA at 1 mg/mL significantly promoted the transcriptional levels (P <  0.01) of Oct-4 and Nanog in hAECs, but only gave rise to a slight increase in Sox-2 expression without significance (Fig. 3A). In agreement with the gene expression, levels of proteins Nanog and Oct-4 were markedly increased after 1 mg/mL HA treatment compared to the control group without HA (Fig. 3B). The result indicates that HA may act to enhance the maintenance of stemness in culture.

We also analyzed the multilineage differentiation capabilities of hAECs after HA treatment. Chemical staining reveals that, after induction, similar to the control, hAECs with HA treatment still generated bright red calcium salt nodules (osteogenic differentiation), purple-blue glycosaminoglycans (chondrogenic differentiation), and red lipid droplets (adipogenic differentiation), respectively (Fig. 3C). These results suggest that the multilineage differentiation capability of hAECs is well retained after HA treatment.

Effects of HA on the paracrine action of hAECs

We then explored the effect of 300 kDa HA on the paracrine of two anti-inflammatory factors secreted by hAECs, IL-10 and TGF-β1 (Skeen et al., 2012), as shown in Fig. 4. After treatment with HA at different concentrations (0.05, 0.1, 0.5, and 1 mg/mL), IL-10 secretion for all groups was significantly increased when compared with the control group (P <  0.05 or P <  0.01). HA at the doses of 0.05, 0.1, and 1 mg/mL also significantly up-regulated the secretion of TGF-β1 when compared with the control group (P <  0.05). We could find that the secretion of IL-10 and TGF-β1 peaked at 6.03 ± 0.07 pg/mL and 5.22 ± 0.17 ng/mL, respectively, in the presence of 1 mg/mL HA. These results indicate that 300 kDa HA may enhance the secretion of anti-inflammatory cytokines in hAECs.

Effects of HA on the cell cycle of hAECs

We also evaluated the effect of 300 kDa HA on the cell cycle of hAECs (Table 2 and Fig. S2). As HA dosage increased, the cell percentage in G0/G1 phase decreased from 88.38 ± 1.18% (0 mg/mL) to 79.69 ± 1.52% (1 mg/mL), while the cell percentage in S phase sharply elevated from 5.84 ± 0.24 (0 mg/mL) to 13.20 ± 1.87 (1 mg/mL) (with an unexpected reduction to 1.20 ± 1.32% at 0.05 mg/mL) (Table 2). Simultaneously, the cell percentage in the G2/M phase rose after HA treatment, but not in a dose-dependent manner. Collectively, we find that as the increase of 300 kDa HA doses, the percentage of cells in S and G2/M phases increase, indicating that 300 kDa HA could regulate the cell cycle transition and drive the cells into a proliferating state.

Effects of HA on senescence of hAECs

We then determined whether HA could ameliorate the senescnece of hAECs. After 48 h of HA treatment, compared with the control, the percentage of SA- β-Gal positive cells significantly decreased from 38.48 ± 1.56% to 5.51 ± 0.77%, suggesting the significant effect of 300 kDa HA on retarding the senescence of hAECs (Fig. 5).

Acting mechanism of HA promoting the proliferation of hAECs

We further investigated the signaling pathway involved in the HA-induced proliferation. In our preliminary screening by PCR microarray, the obvious fold change of signaling pathway genes (SMAD5, BMP3, BMP4, BMP6, BMP7, TGFBR3, and BMPR1B) in hAECs was observed after HA treatment (Table 3). Further validation by RT-qPCR demonstrated that HA significantly up-regulated the expression of SMAD3, SMAD4, BMP4, BMP7, TGFBR3, and BMPR1B (P <  0.05 or P <  0.01), but significantly down-regulated the expression of BMP6 (P <  0.05) (Fig. 6). These results strongly suggest the correlation between the pro-proliferative effect and the TGF-β/BMP signaling pathway.

We further used SB431542, a specific blocker of TGF-β/BMP signaling pathway, to treat hAECs with or without 300 kDa HA exposure. As shown in Fig. 7A, the cell number in the HA group was significantly reduced after addition of SB431542 (P <  0.01). Consistently, the transcriptional levels of proliferation-associated factors Ki67 and PCNA were significantly inhibited after SB431542 treatment (P <  0.01) (Fig. 7B). SB431542 could also significantly suppress the expression of TGF-β/BMP signaling genes, including BMP4, BMP7, TGFBR3, BMPR1B, SMAD3, SMAD4 and SMAD5 (P <  0.05 or P <  0.01) (Fig. 7C). Furthermore, HA treatement could significantly incease the expression of phosphorylated Smad2/3 (p-Smad2/3), while this effect could be reversed by the addition of SB431542 (Fig. 7D). These results strongly suggest that the pro-proliferative effect of 300 kDa HA on hAECs is associated with the TGF-β/BMP signaling pathway.